Information Theory, Learning and Big Data

Mar 16, 2015 to Mar 20, 2015

Click on the titles of individual talks for abstract, slides and archived video.
The workshop is oversubscribed; only registered participants are permitted to attend.
We will be using an overflow room showing the live video feed in order to accommodate as many registered participants as possible. Seating in the auditorium is not guaranteed and will be on a first-come, first-served basis for registered participants only. We strongly suggest that participants arrive early for each talk to secure seating in the auditorium. Doors will open each day at 8:30 a.m.
Talks will also be streamed live on this page during the workshop. If you did not register, we hope that you will be able to either join us for the live webcast or enjoy the videos at a later time.

All events take place in the Calvin Lab Auditorium.

Monday, March 16th, 2015
8:30 am–9:00 am		Coffee and Check-In
9:00 am–9:15 am		Opening Remarks
9:15 am–9:50 am		Introductions
		Sparsity and Base Learning
9:50 am–10:15 am		Low-Rank Matrix Recovery Through Rank-One Projections Tony Cai, University of Pennsylvania
10:15 am–10:40 am		Sketching for M-Estimators: A Unified Approach to Robust Regression David Woodruff, IBM Almaden
10:40 am–11:10 am		Break
11:10 am–11:35 am		Simple, Efficient and Neural Algorithms for Sparse Coding Ankur Moitra, Massachusetts Institute of Technology
11:35 am–12:00 pm		Learning a Hidden Basis through Imperfect Measurements: Why and How Misha Belkin, Ohio State University
12:00 pm–1:30 pm		Lunch
		Clustering and Community Organization
1:30 pm–1:55 pm		Constrained Rank Aggregation and Correlation Clustering Olgica Milenkovic, University of Illinois, Urbana-Champaign
1:55 pm–2:20 pm		Testing Distribution Families Jayadev Acharya, Massachusetts Institute of Technology
2:20 pm–2:50 pm		Break
2:50 pm–3:15 pm		Fundamental Limit of Community Detection in Stochastic Block Models Emmanuel Abbe, Princeton University
3:15 pm–3:40 pm		How Large is the Norm of a Random Matrix? Ramon van Handel, Princeton University
3:40 pm–4:05 pm		Nearly Linear-Time Algorithms for Structured Sparsity Piotr Indyk, Massachusetts Institute of Technology
4:05 pm–4:30 pm		Discussion, Open Problems and Lucky Talk Selection
4:30 pm–6:00 pm		Reception

Tuesday, March 17th, 2015
8:30 am–9:30 am		Coffee and Check-In
9:30 am–9:50 am		Lucky Talk: From Coding to Clustering Emmanuel Abbe, Princeton University
		Distribution Properties
9:50 am–10:15 am		Chebyshev Polynomials, Moment Matching and Optimal Estimation of the Unseen Yihong Wu, University of Illinois, Urbana-Champaign
10:15 am–10:40 am		Information Measure Estimation and Applications: Boosting the Sample Size from n to n log n Jiantao Jiao, Stanford University
10:40 am–11:10 am		Break
11:10 am–11:35 am		Sample Complexity of Estimating Entropy Himanshu Tyagi, Indian Institute of Science
11:35 am–12:00 pm		Minimum Rényi Correlation Principle: From Marginals to Joint Distribution Farzan Farnia, Stanford University
12:00 pm–1:30 pm		Lunch
		Sequential Estimation and Compression
1:30 pm–1:55 pm		Adaptive Coding over Countable Alphabets Stéphane Boucheron, Université Paris Diderot
1:55 pm–2:20 pm		On Learning Distributions from their Samples Sudeep Kamath, Princeton University
2:20 pm–2:50 pm		Break
2:50 pm–3:15 pm		Efficient Minimax Optimal Strategies for Universal Prediction Peter Bartlett, UC Berkeley
3:15 pm–3:40 pm		On the Complexity of Best Arm Identiﬁcation in Multi-Armed Bandit Models Aurélien Garivier, University of Toulouse
3:40 pm–4:00 pm		Discussion, Open Problems and Lucky Talk Selection

Wednesday, March 18th, 2015
8:30 am–9:30 am		Coffee and Check-In
9:30 am–9:50 am		Lucky Talk: Chebyshev Polynomials, Moment Matching and Optimal Estimation of the Unseen Yihong Wu, University of Illinois, Urbana-Champaign
		Information Aggregation and Pattern Matching
9:50 am–10:15 am		Low Regret Recommendations and Item-Item Collaborative Filtering Devavrat Shah, Massachusetts Institute of Technology
10:15 am–10:40 am		Topic Modeling Approach for Rank Aggregation Venkat Saligrama, Boston University
10:40 am–11:10 am		Break
11:10 am–11:35 am		Machine Learning from Human Comparative Judgments Robert Nowak, University of Wisconsin-Madison
11:35 am–12:00 pm		Analytic Pattern Matching: From DNA to Twitter Wojciech Szpankowski, Purdue University
12:00 pm–1:30 pm		Lunch
		Learning and Information Theory
1:30 pm–1:55 pm		Strong Data Processing Inequalities and Estimation with Constraints John Duchi, Stanford University
1:55 pm–2:20 pm		Strong Data Processing Inequalities: Applications to MCMC and Graphical Models Maxim Raginsky, University of Illinois, Urbana-Champaign
2:20 pm–2:50 pm		Break
2:50 pm–3:15 pm		Near-Optimal Hypothesis Testing via Convex Optimization Arkadi Nemirovski, Georgia Institute of Technology
3:15 pm–3:40 pm		Data Driven Convergence in Statistical Estimators Narayana Santhanam, University of Hawaii
3:40 pm–4:00 pm		Discussion, Open Problems and Lucky Talk Selection

Thursday, March 19th, 2015
8:30 am–9:30 am		Coffee and Check-In
9:30 am–9:50 am		Lucky Talk: Low-regret Recommendations and Item-item Collaborative Filtering Guy Bresler, Massachusetts Institute of Technology for Devavrat Shah, Massachusetts Institute of Technology
		Optimization and Learning Models
9:50 am–10:15 am		Confidence-Based Active Learning Kamalika Chaudhuri, UC San Diego
10:15 am–10:40 am		Constructing Informative Features for Discriminative Learning Animashree Anandkumar, UC Irvine
10:40 am–11:05 am		Robust Inference and Local Algorithms Yishay Mansour, Tel Aviv University
11:05 am–11:35 am		Break
11:35 am–12:00 pm		Tradeoffs in Large Scale Learning: Statistical Accuracy vs. Numerical Precision Sham Kakade, Microsoft Research New England
12:00 pm–12:25 pm		Computation-Statistics Tradeoffs in Unsupervised Learning via Data Summarization Mesrob Ohannessian, UC San Diego
12:25 pm–2:00 pm		Lunch
		Learning Structured Distributions
2:00 pm–2:25 pm		Statistical Property Testing and Estimation Beyond the i.i.d. Setting Gregory Valiant, Stanford University
2:25 pm–2:50 pm		Nearest Neighbor Based Greedy Coordinate Descent Pradeep Ravikumar, University of Texas, Austin
2:50 pm–3:20 pm		Break
3:20 pm–3:45 pm		Learning and Testing Structured Distributions Ilias Diakonikolas, University of Edinburgh
3:45 pm–4:10 pm		Approximating Spherical Gaussian Mixtures Ananda Suresh, UC San Diego
4:10 pm–4:30 pm		Discussion, Open Problems and Lucky Talk Selection

Friday, March 20th, 2015
8:30 am–9:30 am		Coffee and Check-In
9:30 am–9:50 am		Lucky Talk: An Algorithmic Characterization of a Class of Inequalities Gregory Valiant, Stanford University
		Computations for Big Data
9:50 am–10:15 am		Detection and Estimation through an Information Theory Lens Sergio Verdú, Princeton University
10:15 am–10:40 am		Testing Probability Distributions using Conditional Samples Rocco Servedio, Columbia University
10:40 am–11:10 am		Break
11:10 am–11:35 am		Sample Starved Correlation Mining Al Hero, University of Michigan
11:35 am–12:00 pm		Fast Approximations of the Pattern Maximum Likelihood Estimate Pascal Vontobel, Chinese University of Hong Kong
12:00 pm–1:30 pm		Lunch
		Related Topics
1:30 pm–1:55 pm		Spy vs Spy: Anonymous Messaging Pramod Viswanath, University of Illinois, Urbana-Champaign
1:55 pm–2:20 pm		Learning Sparse Data with Near-optimal Speed and Efficiency from a Variety of Measurement Processes Sidharth Jaggi, Chinese University of Hong Kong
2:20 pm–2:45 pm		Must One Learn the Channel to Communicate at Capacity? Aaron Wagner, Cornell University
2:45 pm–3:10 pm		Discussion and Open Problems